Cardiomyocytes undergo active cell division with beating autonomously before birth, but immediately after birth they lose the ability to divide, and since they have little undifferentiated progenitor cells, when cardiomyocytes die due to exposure to various forms of stress including myocardial infarction, myocarditis and the like, the lost cardiomyocytes cannot be regenerated. As a result, the surviving cardiomyocytes try to maintain myocardial function through compensatory hypertrophy and the like, but if the stress continues and exceeds an allowable threshold, it leads to further exhaustion and death of cardiomyocytes and a consequent lowering of myocardial function (that is, heart failure).
Therefore, it would seem that methods of transplantation to replace weakened or lost cardiomyocytes would be extremely useful for the treatment of heart failure. In fact, it is known from animal experiments that when immature cardiomyocytes obtained from fetuses are transplanted into adult cardiac tissue, the transplanted cells function effectively as cardiomyocytes (See Non-Patent Document 1). However, it is difficult to obtain sufficient cardiomyocytes for this method, and application to clinical medicine is also difficult from an ethical standpoint.
Attention has therefore focused in recent years on inducing differentiation of stem cells into cardiomyocytes and using these cells for transplantation. At present it has not yet been possible to clearly identify a population of progenitor cells or stem cells capable of producing cardiomyocytes in adult cardiac tissue, so pluripotent stem cells, which are less differentiated and can differentiate into a variety of cells, are considered to be useful for the above method.
Pluripotent stem cells are defined as cells which are capable of indefinite or long-term cell proliferation while remaining in an undifferentiated state in an in vitro culture, which retain normal karyotypes, and which have the ability to differentiate into all of three germ layers (ectoderm, mesoderm and endoderm) under appropriate conditions. At present, the three well-known pluripotent stem cells are embryonic stem cells (ES cells) derived from early-stage embryos, embryonic germ cells (EG cells) derived from primordial germ cells at the embryonic stage, and multipotent adult progenitor cells (MAPC) isolated from adult hone marrow.
In particular, it has long been known that ES cells can be induced to differentiate into cardiomyocytes in vitro. Mouse ES cells were used in most of the early studies. When ES cells are cultured in suspension culture as single cells (individual cells dispersed with no adhesion between cells due to enzyme treatment or the like) without the presence of a differentiation-inhibiting factor such as leukemia inhibitory factor (LIF) or the like, the ES cells adhere to one another and aggregate, forming a structure called embryoid bodies (EBs) which are similar to the early embryonal structures. It is also known that cardiomyocytes with spontaneous beating ability appear when these EBs are cultured in suspension or in adhesion on the surface of culture devices.
ES cell-derived cardiomyocytes prepared as described above exhibit very similar properties to those of immature cardiomyocytes in fetal hearts (See Non-Patent Documents 2 and 3). Moreover, it has been confirmed from animal experiments that when ES cell-derived cardiomyocytes are actually transplanted into adult cardiac tissues, they are highly effective, with results similar to those obtained by transplantation of fetal myocardium (See Patent Document 1; Non-Patent Document 4).
In 1995, Thomson et al. first established ES cells from primates (See Patent Document 2; Non-Patent Document 5), and thus the regeneration therapy using pluripotent stem cell-derived cardiomyocytes has become realistic. Subsequently they also succeeded in isolating and establishing human ES cell lines from early human embryos (See Non-Patent Document 6). Moreover, Gearhart et al. established human EG cell lines from human primordial germ cells (See Non-Patent Document 7; Patent Document 3).
Kehat et al. (See Non-Patent Document 8) and Chunhui et al. (See Patent Document 4; Non-Patent Document 9) have reported that human ES cells can differentiate into cardiomyocytes, as mouse ES cells can do. According to these reports, cardiomyocytes derived from human ES cells not only have the ability to beat spontaneously but also express and produce cardiomyocyte-specific proteins such as myosin heavy and light chains, α-actinin, troponin I and atrial natriuretic peptide (ANP) and cardiomyocyte-specific transcription factors such as GATA-4, Nkx2.5, MEF-2c and the like, and from microanatomical observation and electrophysiological analysis it appears that they retain the properties of immature cardiomyocytes at the fetal stage, and could be used for regenerative therapy.
However, one serious problem remains to be elucidated to use pluripotent stem cell-derived cardiomyocytes for cell transplantation therapy and other purposes. When EBs are formed from ES cells or EG cells by conventional methods, not only cardiomyocytes, but also other types of differentiated cells, such as blood cells, vascular cells, neural cells, intestinal cells, bone and cartilage cells and the like, are developed. Moreover, the proportion of cardiomyocytes in these differentiated cell population is not so high, only about 5% to 20% of the total.
Methods of isolating only cardiomyocytes from a mixture of various kinds of cells include a method of adding an artificial modification to the ES cell genes, conferring drug resistance or ectopic expression, and collecting cells having the properties of cardiomyocytes or progenitor cells thereof. For example, by introducing a gene cassette capable of expressing a neomycin (G418) resistance gene under the control of the α-myosin heavy chain promoter into mouse ES cells, Field and his co-researchers established a system in which those ES cells could only survive in medium to which G418 had been added when they differentiated into cardiomyocytes and expressed the α-myosin heavy chain gene (See Patent Document 1; Non-Patent Document 4). 99% or more of G418-resistant cells selected by this method were confirmed to be cardiomyocytes. However, although the purity of the cardiomyocytes is extremely high in this method, the final number of cardiomyocytes obtained is only a few percent of the total cell count, making it difficult to obtain enough amounts of cardiomyocytes for transplantation.
Recently, Chunhui et al. have reported that when human ES cells are treated with 5-azacytidine, the percentage of troponin I-positive cells (cardiomyocytes) in EBs rises from 15% to 44% (See Non-Patent Document 9), but even in this method the percentage of cardiomyocytes in EBs does not exceed 50%. Moreover, 5-azacytidine is a demethylation agent that alters the expression of genes by removing methyl groups bound to DNA, and because it acts directly on the chromosomes, it is not a suitable drug for preparing cells for cell transplantation.
Other methods for producing cardiomyocytes more efficiently from ES cells include, in the case of mouse ES cells, addition of retinoic acid (See Non-Patent Document 10), ascorbic acid (See Non-Patent Document 11), TGFβ, BMP-2 (See Non-Patent Document 12), PDGF (See Non-Patent Document 13) and Dynorphin B (See Non-Patent Document. 14) and treatment to increase reactive oxygen species (ROS) (See Non-Patent Document 15) and Ca2+ (See Non-Patent Document 16) in the cells, all of which are known to act positively to induce cardiomyocyte differentiation. However, cardiomyocyte-specific or selective differentiation has not been achieved with any of these methods.
The inventors of the present invention have found that when a substance that inhibits bone morphogenic protein (BMP) signaling, particularly Noggin, is added to medium during a certain stage of culture, cells having beating ability which are identified as cardiomyocytes are produced with much higher selectivity and efficiency than in conventional methods (see Patent Document 5).
Granulocyte colony-stimulating factor (hereinafter referred to as G-CSF), which is a hematopoietic factor that was discovered as a differentiation-inducing factor for granulocyte lineage hematopoietic stem cells, is known to promote neutrophil hematopoiesis in the body and hence is clinically used as a therapeutic agent for neutropenia following bone marrow transplantation and/or cancer chemotherapy. In addition to these actions, human G-CSF is reported to act on stem cells to stimulate their differentiation and proliferation, and is also reported to induce recruitment of stem cells in the bone marrow into peripheral blood. It has been reported from in vivo experiments that bone marrow stem cells recruited by G-CSF differentiate into cardiomyocytes in tissue where they were recruited (Patent Document 6; Non-patent Document 17). However, there is no report showing that G-CSF directly induces bone marrow stem cells to differentiate into cardiomyocytes, and there is also no report showing that G-CSF is expressed in embryonic cardiomyocytes or is directly used to induce cardiomyocyte differentiation. Moreover, there is no report showing that G-CSF acts directly on ES cells and is used to induce their differentiation into cardiomyocytes.
As described above, conventional methods alone result in variations in the efficiency of myocardial induction, and there is a demand for a more efficient and selective method for inducing differentiation into cardiomyocytes.
Patent Document 1: U.S. Pat. No. 6,015,671
Patent Document 2: U.S. Pat. No. 5,843,780
Patent Document 3: U.S. Pat. No. 6,090,622
Patent Document 4: International Patent Publication No. WO03/06950
Patent Document 5: International Patent Publication No. WO05/033298
Patent Document 6: International Patent Publication No. WO04/054604
Non-patent Document 1: Soonpaa et al., Science, 264:98, 1994
Non-patent Document 2: Maltsev et al., Mech. Dev., 44:41, 1993
Non-patent Document 3: Maltsev et al., Circ. Res., 75:233, 1994
Non-patent Document 4: Klug et al., J. Clin. Invest., 98:216, 1996
Non-patent Document 5: Thomson et al., Proc. Natl. Acad. Sci. USA, 92:7844, 1995
Non-patent Document 6: Thomson et al., Science, 282:114, 1998
Non-patent Document 7: Sha blott et al., Proc. Natl. Acad. Sci. USA, 95:13726,
Non-patent Document 8: Kehat et al., J. Clin. Invest., 108:407, 2001
Non-patent Document 9: Chunhui et al., Circ. Res., 91:508, 2002
Non-patent Document 10: Wobus et al., J. Mol. Cell. Cardiol., 29:1525, 1997
Non-patent Document 11: Takahashi et al., Circulation, 107:1912, 2003
Non-patent Document 12: Behfar et al., FASEB J., 16:1558, 2002
Non-patent Document 13: Sachinidis et al., Cardiovasc. Res., 58:278, 2003
Non-patent Document 14: Ventura et al., Circ. Res., 92:623, 2003
Non-patent Document 15: Sauer et al., FEBS Lett., 476:218, 2000
Non-patent Document 16: Li et al., J. Cell Biol., 158:103, 2002
Non-patent Document 17: Minatoguchi et al., Circulation, 109:2572, 2004